Errors when a 1.17.4 unit is rebooted
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
juju-core |
Triaged
|
High
|
Unassigned |
Bug Description
I rebooted a unit that was deployed and related using juju-core 1.17.4 and, once it comes back, the unit logs are filled with these:
2014-03-05 09:27:11 INFO juju runner.go:262 worker: start "uniter"
2014-03-05 09:27:11 ERROR juju.worker.uniter runlistener.go:63 failed to listen on unix /var/lib/
t-psql-
2014-03-05 09:27:11 ERROR juju runner.go:220 worker: exited "uniter": failed to initialize uniter for "unit-psql-1": listen unix /var/lib/
ket: address already in use
2014-03-05 09:27:11 INFO juju runner.go:254 worker: restarting "uniter" in 3s
2014-03-05 09:27:14 INFO juju runner.go:262 worker: start "uniter"
2014-03-05 09:27:14 ERROR juju.worker.uniter runlistener.go:63 failed to listen on unix /var/lib/
t-psql-
2014-03-05 09:27:14 ERROR juju runner.go:220 worker: exited "uniter": failed to initialize uniter for "unit-psql-1": listen unix /var/lib/
ket: address already in use
2014-03-05 09:27:14 INFO juju runner.go:254 worker: restarting "uniter" in 3s
Full logs attached.
tags: | added: reboot regression |
Changed in juju-core: | |
status: | New → Triaged |
importance: | Undecided → High |
milestone: | none → 1.18.0 |
confirmed. I just experienced this on a test system.
It appears to be related to the recent additions of the juju run
support. It creates a UNIX socket to communicate on. While starting
up, it tries to create it, but since it already exists it fails, but
will never actually succeed.
We probably should just be using one that is there, or deleting it if
we start up and find one already there.
On Wed, Mar 5, 2014 at 1:09 PM, Curtis Hovey <email address hidden> wrote: /bugs.launchpad .net/bugs/ 1288141 /bugs.launchpad .net/juju- core/+bug/ 1288141/ +subscriptions
> ** Tags added: reboot regression
>
> ** Changed in: juju-core
> Status: New => Triaged
>
> ** Changed in: juju-core
> Importance: Undecided => High
>
> ** Changed in: juju-core
> Milestone: None => 1.18.0
>
> --
> You received this bug notification because you are subscribed to juju-
> core.
> https:/
>
> Title:
> Errors when a 1.17.4 unit is rebooted
>
> To manage notifications about this bug go to:
> https:/